1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

BIOS Freezes before POST completes - high end PC problem.

Discussion in 'General Hardware' started by RuskiSnajper, Jan 28, 2011.

  1. RuskiSnajper

    RuskiSnajper

    Joined:
    Feb 18, 2009
    Messages:
    1,608 (0.77/day)
    Thanks Received:
    231
    Location:
    slovenia , europe
    My specs should be clear. (right side)

    I have AWARD bios as you can see , and recently i have lockups at boot , not re-boot , but only a boot from power off. (hard-(re)boot)

    It detects my CPU and MOBO and BIOS version , but it doesn't go on , HDD stops (if it even starts?) , Keybord doesn't even lit up , and the system is not responding in bios (pre-post)


    When it's going to fail is before "memory testing" , if it doesn't display that like 0.5 secs after detecting CPU and Mobo then it will lock up , I have no idea what the heck is going around the PC is super stable as i fixed and upgraded ram and it works fine everything is spotless , but i have not tested with memtest since June 2010 (latest format) i think that's fairly early so ... can't be breaking something that fast.


    Maybe something got damaged ? bios chip it self ? there are no beep errors , i think , i think only one beep before it even finishes post , yes i have one beep always when the PC starts POST. (first screen)


    If it displays "Memory testing XXXXXXX" then it will do the rest and CMOS was never erased. I didn't tried flashing or CLR_CMOS , the last time i flashed was like early 2010 .. the time when F14 bios version came out for my mobo.



    However , i recall now , something , interesting , This probably started after i installed the stupid Ad-Aware from Lavasoft , this program is totally broken as far as im concerned , i had to manuall delete bootexecutes about some "lsdelete autocheck" crap that started before windows. The program wasn't what it used to be , now it finds nothing , it's a stupid UI , it lags on a high end machine , and it has 20 of it's processes opened , it's like a freaking virus.

    The only thing currently to boot is for me to reset the PC multiple times , i had to do it just twice or three times (hard-reboot doesn't help) , i took the PC for cleaning today (that includes taking it downstairs and out to the yard and blowing it with an industrial 400volt air compressor :cool:) - it was cold today , like -2 Celsius , (apart from last week) , but i keept it to get used to room temperature for like an hour when i finished , i also kept sure that i din't spin the fans to much as it was noted it works like a dinamo (generating charge and going reverse) and makes electricity which would be harmful to components.




    ---- I don't really know what to listen , the 3DGAMEMAN says it's good to keep the POWER PLUG inside the PSU when working with PC , but the manuals from gigabyte says always disconnect and power off anything completely , 3DGM was referring to the grounding of the PSU (EU: the extra 2 bended wires seen around the plug) - i don't reallly know what to do anyways , is it better to have it plugged in when spinning the fans manually so the charge would go into the psu and ground before going to the PC components ?


    That actually didn't cause the bios freezing problem , it couldn't , this started happening like amonth ago and i didn't cleaned it from the last time i finished setting up the new format (september 2010)
     
    Last edited: Jan 28, 2011
  2. theoneandonlymrk

    theoneandonlymrk

    Joined:
    Mar 10, 2010
    Messages:
    3,411 (2.02/day)
    Thanks Received:
    572
    Location:
    Manchester uk
    sounds like your memmory is on its way out try single stick or diff memmory thats first to try also could be your power supply extremely long post dude i had to stop for a brew:)

    psu plugged in but switched hard switched at back off then its not live but grounded tho dont be opening the psu up it will have a few live bits
     
    jsfitz54 says thanks.
    More than 25k PPD
  3. RuskiSnajper

    RuskiSnajper

    Joined:
    Feb 18, 2009
    Messages:
    1,608 (0.77/day)
    Thanks Received:
    231
    Location:
    slovenia , europe
    I think my PSU Discharges completely when i turn it off , since i don't have the momentairly turn up unlike another PC has who does not have enermax PSU. ... i guess that's a feature.

    But these are top line corsair dominators , i can't believe they could be dying from less than a year of use , the 2x2 GB are mere a few months old , while 2x1GB are like 10 months. I did suspect that of course but you know i was trying to hope that it wasn't , no please no!
     
    Last edited: Jan 28, 2011
  4. Swamp Monster

    Swamp Monster

    Joined:
    Oct 22, 2010
    Messages:
    561 (0.38/day)
    Thanks Received:
    84
    Location:
    Latvia
    +1 If you have switch at back, switch it off, but don't unplug the cord. Then it will still be grounded.


    Charge from manually spinning fans could still damage components. It's better to unplug all fans before cleaning with compressor.
     
    jsfitz54 says thanks.
  5. RuskiSnajper

    RuskiSnajper

    Joined:
    Feb 18, 2009
    Messages:
    1,608 (0.77/day)
    Thanks Received:
    231
    Location:
    slovenia , europe
    OH no

    My Dominators got broken , the problem is so small , so tiny one bit , but the error count is like above 60 and i only got 44% throughout the memtest

    The problem is not affecting the machine when running or gaming, these rams are top DDR2 rams but i guess , it's just a mix of bad settings and all but i don't think so everthing was wrong,

    It's a relief the MOBO isn't to blame it's 3 years old and i don't have a warranty for that , but i do have for all the ram.

    The EPP1 is different than the information shown on the sticker


    (im running them on 800 but they are 1066 certified , i was having automatic clocks , and manual voltage control set to normal , they were on 5-5-5-18 i think )


    [​IMG]


    Total 6 GB , the error is fairly unnoticable in gaming as i sad , nothing happens , it's just when i BOOT the computer , it seems like it needs some time to "warm up" before i get a pass , i rebooted it 3 times yesterday , i got a BIOS CHECKSUM ERROR , so basically it downgraded/backuped to F9 bios version (but the settings remained the same)

    Today , i rebooted it 10 times before i got a pass , when i finnaly got into the bios i started memtest and it was a flood of errors at the start , all happening in a super small bit range throughout the whole test , about more than 62 of them. ex. " Err-Bits 0000000001 " , i didn't even test to 100% , i will revise settings i hope i get them right but i don't see a problem other than small overvoltage becasue the STUPID BIOS doesn't provide NEGATIVE voltage change ; WHY ????? , Why am i forced to set lower clocks when lowering frequency with lowest voltage.

    The address was limited to only one , so THANK GOD ONLY ONE STICK , but i don't have idea which stick yet , the 1 GB or 2 GB stick ? , :(


    Basically i set it to JEDEC 2 , but i manualized everything , i think i don't know , or didn't knew how to get automated JEDEC/EPP settings out of bios , i think i realized , i need to set manual frequency , and then put everything else on auto and it will set for which frequency ends up matching JEDEC/EPP , but it wasn't set like it. "System voltage not optimized* was blinking as usual , but i disregarded it because i wanted to set loadline calibration to disabled since you can't set that if you have set voltage control on Auto , no F14 BIOS is the newest but it doesn't fix this.

    The F9 bios is old and the only problem with it is the CPU FINE MULTIPLIER bug where it doesn't register the +.5 ratio so my CPU frequency is lower than stock at 2.33 instead of 2.50mhz , that's not even a problem get's fixed after bios update.

    I am going to test each stick now , ill get back when i identify it and i will test a working one to see bios boot behavitor , i REALLY hope it's not the mobo failing, that's probably the user's worst fears.

    But you know what , if i had Sys Voltage on MANUAL , and every sub option on NORMAL , it was the same voltage for ram and other stuff as having AUTO , so ... really , that's why i disregarded the flashing warning. Still i can't find anything really wrong with my settings.

    I can't believe these dominators can be broken so quicky , but i guess this speaks quality ,the error were a lot but it was so small address that it doesn't make a difference in operationability/stability ... it's just the bios that has the problem with it , weird , brother had a pair of 1GB good working Crucial's also top line performance and quality , but for some reason they DIED so hard one day (over-the-night) they can't be tested, they can't be ran , the (other one) PC will not boot with them anymore.
     
    Last edited: Feb 6, 2011
  6. jsfitz54

    jsfitz54

    Joined:
    Jun 18, 2010
    Messages:
    903 (0.57/day)
    Thanks Received:
    242
    Use the most recent Bios and run only one memory bank/stick at a time while running memtest.

    See: theoneandonlymrk advice above. Of course you don't know which stick is bad.
     
  7. RuskiSnajper

    RuskiSnajper

    Joined:
    Feb 18, 2009
    Messages:
    1,608 (0.77/day)
    Thanks Received:
    231
    Location:
    slovenia , europe
    A weirdness is confirmed:

    The ram needs 2 minute warm up before it can pass bios test , reboots don't matter , sleep/S3 standby works because the ram is powered an thus warmed up.

    Is this showing the awesome resilience of dominators ? Even if they fail , they don't fail like other's , for example the old ram had 2 MB failed address range , how can this be 1 bit , priceless.

    I can still work with the PC , everything , same as before, but im definitely replacing them cause i still have warranty.

    Also , nothing's 100% superworking , it's an interesting thing but still everything can fail , it's just impossible to make 100% working, so , this is just not enough to say corsair is bad. not at all.
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page